PyLayers
A Site-Specific Mobile Indoor Radio Channel Modeling Platform
From Graph-Based Ray Tracing to Indoor Positioning
Overview
2014
What is PyLayers ?¶
- PyLayers is Open Source
- PyLayers is an evolutive platform for education and research in radio propagation
- PyLayers is a radio propagation simulator fully written in Python langage
- PyLayers is organized as a set of specialized independent modules
- Good for maintainability and further evolution
- PyLayers adresses the prediction of radio observable for localization in Indoor environments
- PyLayers adresses various radio standard including UWB radio
- PyLayers adresses realistic human mobility and motion
- PyLayers aims to connect with upper Layer abstraction
List of the main modules
Layout of Indoor Environments
- The very first thing to do, is to describe the selected propagation environement.
- The manner this environement is described has very strong implications on rays algoritms performances.
- The design choice which has been made, is to heavily relying on the graph data structure
To go further, we need to import the following specialized module
In [2]:
from pylayers.gis.layout import *
Loading a Floor Plan (Layout)
- The chosen environement corresponds to the office building where an UWB Impulse Radio measurement campaign took place during the WHERE1 FP7 project.
- Several format are supported (including
.osm)
In [3]:
L=Layout('WHERE1.ini')
f = figure(figsize(30,30))
plt.axis('off')
f,a = L.showG('r',nodes=True,node_size=80,node_color='g',edge_color='b',width=5,fig=f)
t = plt.title('Graph of rooms',fontsize=40)
Graph of Ways
In [4]:
f = figure(figsize(30,30))
plt.axis('off')
f,a = L.showG('w',nodes=True,node_size=80,node_color='c',edge_color='k',width=1,fig=f)
t =plt.title('Graph of Ways',fontsize=40)
Link Simulation
For enabling electromagnetic simulation we load the following modules
radionode- coordinates
- 3D orientation
- antenna object
link- 2
RadioNode - 1
Layout
- 2
In [5]:
from pylayers.simul.radionode import *
from pylayers.simul.link import *
Loading the Tx and Rx points from the WHERE1-M1 measurement campaign.
In [6]:
rx = RadioNode(_fileini='w2m1rx.ini',_fileant='defant.vsh3')
tx = RadioNode(_fileini='w2m1tx.ini',_fileant='defant.vsh3')
Link Simulation
- Links are saved in
hdf5format - All Links data members are evaluated once and stored
In [18]:
#Lk = DLink(L=L,a=tx.points[1],b=rx.points[54],cutoff=5,verbose=False)
Lk = DLink(L=L,a=tx.points[2],b=rx.points[114],cutoff=5,verbose=False)
Lk.eval()
f,a = Lk.show(labels=False)
Main Link members
The Signature .Si
In [19]:
Lk.Si.keys()
print Lk.Si[12].shape
print Lk.Si[12][0,:]
print Lk.Si[12][1,:]
- Signatures are rays potentialities
- A set of Signature is calculated from one Layout cycle to an other Layout cycle
- A Signature consists of 2 lists
- A list of interaction entities (point (<0) or segment (>0))
- A list of type of interactions (1:R, 2:T, 3:D)
- The same signature is used for calculation of rays from one cycle to an other cycle
Main Link Members
The Rays .R
In [20]:
Lk.R
Out[20]:
Group of rays with 3 interactions
In [21]:
c = plt.cm.jet
col = c(1.0)
f,a = Lk.R.show(L=Lk.L,figsize=(20,20),i=3,alpharay=0.3,widthray=5,colray=col)
Group of rays with 4 interactions
In [22]:
f,a = Lk.R.show(L=Lk.L,figsize=(30,30),i=4,widthray=2,colray='m')
All group of rays (from 3 to 9 interactions)
In [23]:
f,a=Lk.show(rays=True,col='cmap',pol='tot',dB=True,cmap=cm.hot,dyn=50,labels=False)
f,a=Lk.show(rays=True,col='cmap',pol='tot',dB=True,cmap=cm.hot,dyn=120,labels=False)
In [24]:
f,a=Lk.show(rays=True,col='cmap',pol='tt',dB=True,cmap=cm.hot,dyn=50,labels=False)
f,a=Lk.show(rays=True,col='cmap',pol='pp',dB=True,cmap=cm.hot,dyn=50,labels=False)
In [25]:
Ett,Etp,Epp,Ept = Lk.C.energy()
Etot = Ett+Etp+Epp+Ept
semilogy(Etot)
Out[25]:
In [26]:
Lk.R[4]['rayidx']
Out[26]:
In [27]:
clm = cm.copper
f,a = Lk.show()
#rk = Lk.R.keys()[::-1]
rk = Lk.R.keys()
for i in rk:
lr = Lk.R[i]['rayidx']
for r in range(len(lr)):
ir = lr[r]
RayEnergy=max((10*log(Etot[ir]/Etot.max())+40),0)/40
col = clm(RayEnergy)
f,a = Lk.R.show(i=i,r=r,colray=col,widthray=RayEnergy,alpharay=RayEnergy,fig=f,ax=a,layout=False,points=False)
Main Link members
The Propagation Channel .C
- \(\mathbf{\tilde{C}(f)} = \left[\matrix { C_{\theta,\theta} & C_{\theta,\phi}\\C_{\phi,\theta} & C_{\phi,\phi}}\right]\)
In [48]:
Lk.C
Out[48]:
The propagation channel
- Full polarization
- Full frequency
- Full angular information
- At that stage there is no antenna involved
- Vertical axis is the ray index
In [28]:
Lk.C.sort('energy')
f,a= Lk.C.show(cmap='jet',fontsize=40,typ='l20',vmin=-120,vmax=-30)
In [29]:
Lk.C.sort('tauk')
f,a=Lk.C.show(cmap='jet',fontsize=40,typ='l20',vmax=-30,vmin=-120)
In [30]:
Lk.C.sort('tauk')
f,a=Lk.C.show(cmap='jet',fontsize=40,typ='r')
It is possible to truncate the propagation at a given energy threshold. Warning, this operation is not reversible.
In [52]:
#Lk.C.cut(threshold=0.99)
#Lk.C.sort('tauk')
#f,a=Lk.C.show(cmap='jet',fontsize=40,typ='l20',vmax=-30,vmin=-120)
To vizualize an individual ray: + i : group of rays + r : index of the ray in the group of rays
In [31]:
Lk.R
Out[31]:
In [32]:
Lk.R.show(L=Lk.L,i=4,r=-1,colray='green',widthray=1,figsize=(20,20))
Out[32]:
Main Link members
The transmission channel .T
In [33]:
f,la=Lk.H.show(cmap='jet',fontsize=40)
Internal Description of Complex Signals
Materials and Slabs
Antennas and Propagation
Simulation of a Radio Link
Handling human mobility
Comparison with measurements
Thanks!
In [57]:
from IPython.display import display, HTML
s = """
<style>
.rendered_html {
font-family: "proxima-nova", helvetica;
font-size: 150%;
line-height: 1.3;
}
.rendered_html h1 {
margin: 0.25em 0em 0.5em;
color: #015C9C;
text-align: center;
line-height: 1.2;
page-break-before: always;
}
.rendered_html h2 {
margin: 1.1em 0em 0.5em;
color: #26465D;
line-height: 1.2;
}
.rendered_html h3 {
margin: 1.1em 0em 0.5em;
color: #002845;
line-height: 1.2;
}
.rendered_html li {
line-height: 1.5;
}
.prompt {
font-size: 120%;
}
.CodeMirror-lines {
font-size: 120%;
}
.output_area {
font-size: 120%;
}
notebook {
background-image: url('files/images/background.png');
}
h1.bigtitle {
margin: 4cm 1cm 4cm 1cm;
font-size: 300%;
}
h3.point {
font-size: 200%;
text-align: center;
margin: 2em 0em 2em 0em;
#26465D
}
.logo {
margin: 20px 0 20px 0;
}
a.anchor-link {
display: none;
}
h1.title {
font-size: 250%;
}
</style>
"""
display(HTML(s))